Hyperspace Hashing is a new technique for determining how to distribute objects onto nodes in a cluster. Specifically, it maps objects with multiple attributes into points in a multidimensional hyperspace, which has in turn been divided into zones assigned to servers. Search queries on secondary object attributes can therefore be mapped to small, hyperspace regions representing the set of feasible locations for the matching objects. This geometric mapping enables efficient searches that do not require enumerating across every object in the system.

Concepts — HyperDex 0.2b9 documentation

The homepage.

Similar to redis but faster and more flexible.

The quote above makes me think I've found home.